CTE DB SUA 8 in Spain
Why a Spanish building project answers the protection question with a code formula rather than an IEC 62305-2 risk assessment, what that formula actually is, and where the international standard does and does not appear.
On a Spanish building project, whether lightning protection is required is decided by a formula in the building code, not by an IEC 62305-2 risk assessment. The two are different calculations, with different inputs, and IEC 62305 is not cited anywhere in the code document.
That is unusual. Most countries adopt IEC 62305 and then argue about parameters at the margins, which is the pattern the rest of these country pages describe. Spain runs a parallel method inside its own mandatory building code, and an engineer arriving from an IEC background will look for a reference to the international standard and not find one.
This page is written from the consolidated Documento Basico SUA published by the Ministerio de Transportes, Movilidad y Agenda Urbana, dated 14 June 2022, whose articulado was approved by Real Decreto 314/2006 and subsequently modified, most recently by Real Decreto 450/2022. Every claim below names the clause, equation, table or annex it comes from. Note that the consolidated version states of itself that it has no legal force, since it is assembled from the successive modifications published in the Boletin Oficial del Estado; the individual dispositions are the legally operative texts. Lumex is independent of the Ministerio and the IEC, and this page does not reproduce the code.
Where SUA 8 sits in the Spanish code
The Codigo Tecnico de la Edificacion is Spain's building code, organised into Documentos Basicos, each serving one basic requirement. DB SUA covers Seguridad de utilizacion y accesibilidad, safety in use and accessibility, and runs from SUA 1, falls, through to SUA 9, accessibility. Lightning sits at SUA 8, Seguridad frente al riesgo causado por la accion del rayo.
The basic requirement itself is stated in article 12.8 of Part I, and it is short: the risk of electrocution and of fire caused by the action of lightning shall be limited, by means of adequate installations of protection against lightning. Note what the requirement names as the two harms. Electrocution and fire. There is no counterpart to the international standard's separate treatment of loss of service to the public or loss of cultural heritage, and no counterpart to the failure of internal electrical and electronic systems that drives so much of a modern IEC assessment.
The section that delivers that requirement has just two clauses: clause 1, the verification procedure, and clause 2, the type of installation required. The characteristics of the installation itself are pushed out to Anejo B. That is the whole of Spain's mandatory building code treatment of lightning, and its compactness is the first clue that it is not a translation of anything in the IEC series.
Two numbers, one comparison
Clause 1 requires an installation when the expected frequency of impacts exceeds the admissible risk. Both sides are computed from the code's own map and tables.
Ne, the expected frequency of impacts
Ne = Ng Ae C1 × 10-6, in impacts per year. Ng is the density of impacts on the ground in impacts per year per square kilometre, read from the contour map at figure 1.1 covering peninsular Spain, the Balearics, the Canaries, Ceuta and Melilla. Ae is the equivalent capture area. C1 is the surroundings coefficient from table 1.1.
Na, the admissible risk
Na = 5,5 ÷ (C2 C3 C4 C5) × 10-3. The numerator is a fixed constant and the four coefficients cover construction type, contents, use and the need for continuity of the activity. There is no judgement call and no investigation to perform: the threshold is produced by the same kind of table lookup as the frequency it is compared against.
Ae is worth reading carefully, because it is defined more precisely than a quick reading suggests. It is the equivalent capture area of the isolated building in square metres, bounded by a line traced at a distance of 3H from each of the points of the perimeter of the building, H being the height of the building at the perimeter point being considered. Because H is taken per perimeter point rather than as one building height, a stepped or terraced building produces a boundary that follows its profile, and taking the maximum height around the whole perimeter overstates the area and therefore the computed frequency.
What the five tables actually weigh
Four coefficients push the admissible risk down, making protection more likely to be required, and one adjusts the frequency. Read together they show what the Spanish code considers to be at stake.
The surroundings
Table 1.1. Close to other buildings or trees of the same height or taller, 0,5. Surrounded by lower buildings, 0,75. Isolated, 1. Isolated on a hill or promontory, 2. This is the only coefficient acting on Ne, and it spans a factor of four between a sheltered site and an exposed one.
The construction
Table 1.2, a grid of structure against roof. A metal structure with a metal roof gives 0,5, and a timber structure with a timber roof gives 3, with concrete combinations sitting between them. A conducting building is treated as partly protecting itself, and a combustible one as the opposite.
The contents
Table 1.3, and it is binary. A building with flammable contents scores 3, and anything else scores 1. There is no gradation between them, which is a marked contrast with the international standard's treatment of fire risk and the reduction factors it allows for fire provisions.
The use
Table 1.4. Buildings not normally occupied, 0,5. The uses Publica Concurrencia, Sanitario, Comercial and Docente, meaning public assembly, healthcare, retail and educational, all score 3. Everything else scores 1. Occupancy by the public multiplies the requirement threefold in one step.
Continuity of service
Table 1.5, also binary and the sharpest of them. Buildings whose deterioration could interrupt an indispensable service, with hospitals and fire stations given as the examples, or could cause a serious environmental impact, score 5. Everything else scores 1.
How far the threshold moves
Because the four coefficients multiply, they compound. A hospital in a timber-framed building with flammable contents drives Na down by the product of its coefficients, while a normally unoccupied metal-framed shed with inert contents pushes it up. The admissible risk is not a constant with adjustments; it is a computed figure that ranges widely.
From efficiency to protection level
Clause 2 converts the comparison into a required system. This is where the two most commonly misread points in SUA 8 live.
E = 1 − Na / Ne
The required efficiency of the installation is one minus the ratio of admissible risk to expected frequency. The further Ne exceeds Na, the closer E approaches 1, and the more capable the system has to be. When Ne does not exceed Na, E is zero or negative and no installation is required.
The level runs backwards
E of 0,98 or above requires level 1. From 0,95 to below 0,98, level 2. From 0,80 to below 0,95, level 3. From 0 to below 0,80, level 4, with a footnote recording that within those limits the installation is not obligatory. The highest demand carries the lowest level number, which catches people out on a first reading.
The second point is the one to carry into a design review. Level 4 in table 2.1 is the row where nothing is required. A note recording that a building came out at level 4 therefore means the code did not require an installation, not that a level 4 system should be specified. That reading error runs in the safe direction, but it produces installations the code did not ask for, and it makes a schedule of buildings hard to interpret. Where the level was genuinely computed, the record should carry Ne, Na and E rather than the level alone, so a reviewer can see which side of the comparison the verdict came from.
Where the calculation does not get a vote
Clause 1.2 removes two categories of building from the procedure entirely. Buildings in which toxic, radioactive, highly flammable or explosive substances are handled, and buildings whose height exceeds 43 m, shall always have lightning protection systems of efficiency E equal to or greater than 0,98.
Reading that against table 2.1, an efficiency of 0,98 or above is protection level 1, the most demanding of the four. So for these buildings the answer is fixed before any site data is gathered, and the design question moves straight to how level 1 is achieved. There is no path by which a favourable Ng, a sheltered C1 or a conducting structure argues the requirement down.
The 43 m threshold is worth knowing precisely, because it is an unusual number and it is absolute. It is not a rounded 45, and it is not tied to a number of storeys. On a project near that height it is a figure to check against the design rather than estimate, since crossing it changes the required level in one step.
The installation, including the part IEC 62305 does not recognise
Once SUA 8 has established that a system is required and at what level, Anejo B specifies what that system is. It must comprise an external system, an internal system and an earth network. Much of it will look familiar: the external system uses air termination devices and down conductors, and the protected volume can be determined by the protection angle, the rolling sphere or the mesh method, used separately or in combination.
The numbers are recognisably of the same family as the international ones. Table B.2 gives the rolling sphere radius by level as 20, 30, 45 and 60 m for levels 1 to 4. Table B.3 gives the mesh size as 5, 10, 15 and 20 m. Table B.5 gives the spacing between down conductors as 10, 15, 20 and 25 m. Anejo B also requires at least one down conductor per air termination and a minimum of two where the horizontal projection of the conductor exceeds its vertical projection or where the structure is taller than 28 m, with equipotential connections between down conductors at ground level and every 20 m. For a building over 60 m protected by mesh, a conductive mesh must additionally protect the upper 20 per cent of the facade.
Then comes the genuine divergence. Anejo B lists the permitted air termination devices as Franklin rods, conductive mesh, and pararrayos con dispositivo de cebado, the Spanish term for a triggered or early streamer emission terminal. Clause B.1.1.2 does not merely permit them; it dimensions them. Below a horizontal plane 5 m beneath the tip, the protected volume is a sphere of radius D plus L, where D comes from table B.4 by protection level and L is a distance in metres derived from the terminal's triggering advance time and capped at 60 m. Above that plane the protected volume is a cone.
That is a real difference from the international standard rather than a difference of wording, and it runs the opposite way to several other jurisdictions. IEC 62305 specifies conventional air terminations and provides no enhanced attraction radius, and Singapore's committee declined to recommend such methods at all. Spain's mandatory building code gives them a geometry. If you work across those markets, read ESE air terminals and IEC 62305 for what the international standard does and does not say, and do not carry a Spanish protected volume onto an IEC job or the reverse.
IEC 62305 is not cited in the Documento Basico
This is a claim worth stating precisely, because a negative is easy to assert loosely and hard to rely on. A full text search of the consolidated Documento Basico SUA, across all of its pages, returns no occurrence of IEC 62305, and none of UNE-EN 62305. Not in Section SUA 8, not in Anejo B which specifies the installation, and not in Anejo C.
Anejo C is the decisive one, because its stated purpose is to collect the complete reference of the standards cited in the articulado of the DB SUA. It lists five: UNE 41901 on slipperiness, UNE-EN 12046-2 on door operating forces, UNE-EN 12600 on glass for building, UNE-EN 81-70 on lift accessibility and UNE 41501 on the accessibility symbol. There is no lightning standard in the list, because Section SUA 8 and Anejo B do not cite one. They are self-contained.
What that does and does not mean is worth separating. It means the CTE's own route to compliance runs entirely on the CTE's own method, so an IEC 62305-2 result is not what the Documento Basico asks for. It does not mean the international standard is unknown or unusable in Spain: UNE-EN IEC 62305 exists as the Spanish adoption of the European standard, Spanish engineers work to it, and international project specifications routinely name it. The code simply does not route through it. And the CTE does provide a general path for other solutions, stating that solutions different from those contained in the DB may be used provided the procedure established in article 5 of the CTE is followed and compliance with the basic requirements is documented in the project. Whether a given alternative satisfies that procedure is a matter for the project and its authority, and not something this page can answer for you.
A computed threshold, not an established one
The deepest difference between the two methods is not a coefficient. It is what kind of thing the threshold is.
In the international standard the threshold is something to establish. Clause 7.3 NOTE 1 gives RT = 1×10-5 per year as a representative value of tolerable risk and adds that another value may be set once the case has been investigated in detail. Printed p.12 then lets national or local regulations fix RT, the tolerable frequency of damage FT, and the Annex A, B, C and E calculation rules and parameter values. The tolerable value is therefore a decision to be made and recorded for a project, from the code, the authority having jurisdiction and the specification that govern it.
The CTE does the opposite. Na is produced by equation (1.2) from a fixed numerator of 5,5 and four table lookups, so the threshold is computed rather than chosen. Spain has, in effect, already made the decision the international standard leaves open, and encoded it in the four coefficients. A Spanish project under SUA 8 does not select a tolerable value, and there is no equivalent of the detailed investigation the IEC text contemplates.
That has a consequence for anyone reconciling the two on one estate. The Spanish figures and the IEC figures are not two estimates of the same quantity that ought to agree. They are outputs of two different models with different scopes, and a discrepancy between them is not evidence that one of them is wrong. State which method produced each figure, and do not average them.
What this means on a Spanish project
Four consequences that follow from everything above, in the order they usually matter.
Answer the code question first
For a building under the CTE, the compliance question is the SUA 8 comparison of Ne against Na. An IEC 62305-2 assessment, however thorough, is not the calculation the Documento Basico asks for, and offering one in its place leaves the code question unanswered.
Record all three numbers
Ne, Na and E, with the coefficient values and the Ng read from figure 1.1. A bare protection level cannot be rechecked, and given that level 4 is the not-required row, a level on its own is ambiguous about whether anything was needed at all.
Check 43 m and the substances
Both overrides bypass the calculation and land straight on level 1. On a building near the height threshold, or on any project handling toxic, radioactive, highly flammable or explosive substances, establish that before spending effort on a calculation that cannot change the answer.
Keep the two methods apart
A multinational client with Spanish and IEC-region sites is running two models, not one with local settings. Say which produced each result, and never move a protected volume derived from a triggered air terminal under Anejo B onto a site governed by IEC 62305.
Where to go from here
Spain is the sharpest departure among the country pages, so the contrast is the useful read. For the international picture, start with IEC 62305 around the world. For a national adoption that keeps the IEC method and changes the parameters, which is the far more common pattern, read BS EN 62305 in the UK or SS 555 in Singapore.
To see what the method Spain does not use actually does, read the IEC 62305-2 risk assessment and how an IEC 62305 risk is calculated. The protection levels and the rolling sphere radii that Anejo B shares with the international standard are explained in lightning protection levels and air-termination methods.
Being clear about what this tool does not do
Lumex computes the IEC 62305-2:2024 method, on the current edition, with every figure traceable to the clause, equation or table behind it. It does not compute CTE DB SUA 8. It does not evaluate Ne and Na, it does not read Ng from figure 1.1, and it does not produce the efficiency E or the protection level that a Spanish building code submission is judged on. Those are a different calculation, and a tool that ran the IEC method and presented the result as CTE compliance would be misrepresenting what it had done. Where Lumex is useful on a Spanish project is where the international standard is genuinely the basis: an international specification naming IEC 62305, a client standardising an estate across borders, or an assessment that has to sit alongside the code check rather than replace it. See the Lumex platform.
Questions answered
What is CTE DB SUA 8?
Does Spain use IEC 62305 to decide whether a building needs lightning protection?
How does CTE DB SUA 8 decide if lightning protection is required?
How is the equivalent capture area Ae defined in the CTE?
Which buildings always need lightning protection in Spain?
What is the efficiency E in CTE DB SUA 8?
Does Spain permit early streamer emission air terminals?
Is the tolerable risk in Spain the same as R_T in IEC 62305?
Does CTE DB SUA 8 apply to every structure in Spain?
Lumex computes the IEC 62305-2 method and shows the working. It does not certify a structure. You may not issue or submit a Lumex output until a competent person, qualified where the structure is located, has reviewed the inputs and the result and signed it.
The tolerable risk in IEC 62305-2 is not a fixed constant. Clause 7.3 NOTE 1 gives RT = 1×10-5 per year as a representative value of tolerable risk and adds that another value may be set once the case has been investigated in detail. Printed p.12 then lets national or local regulations fix RT, the tolerable frequency of damage FT, and the Annex A, B, C and E calculation rules and parameter values. Every Lumex assessment states the jurisdiction it was computed under and the values that applied.